Locke's Philosophy of Science

First published Fri Jul 24, 2009; substantive revision Thu Jul 25, 2013

Locke's philosophy of science consists largely in his metaphysical and epistemological views of material substances and their powers. Locke has been widely hailed for providing an epistemological foundation for the experimental science of his day, and his thought is closely aligned with that of its practitioners, elaborating certain themes present in sparer form in Boyle and Newton. But if his epistemology helps to usher in the age of science, he still belongs to the age of natural philosophy. And if he is a devotee of the new science, he often appears an uncertain one, recognizing profound difficulties in it. In consequence, Locke's work is characterized by tensions and nuances, providing a rich source for scholarly research and debate.


1. Introduction

Two features of Locke's intellectual landscape are most salient for understanding his philosophy of science. The first is the profound shift underway in disciplinary boundaries, in methodological approaches to understanding the natural world, and in conceptions of induction and scientific knowledge. Locke's reaction is in one sense conservative. He retains as an ideal the notion that scientific knowledge is demonstrative and certain, an ideal he shares with the two main targets of his Essay, the Aristotelians and the Cartesians. Yet in another, ultimately more important sense, his reaction is progressive. Impressed by empirical methods and cognizant of their poor fit with the Aristotelian ideal, he defines a distinct kind of knowledge, one inferior to genuine scientific knowledge but appropriate to human sensory capacities. In so doing, he develops an epistemological basis for the new, experimental philosophy.

The second salient feature is the dominant scientific theory of his day, the corpuscular hypothesis. As defined for the purposes of this article, the corpuscular hypothesis (i) takes observable bodies to be composed of material particles or corpuscles, (ii) takes impulse (action by surface impact) to be the primary if not the sole means of communicating motion, and (iii) attempts to reduce qualities at the level of observable bodies, such as color, to the primary, that is, inherent properties of the particles composing those observable bodies. In what may be called its orthodox version (“pure mechanism”, as Ayers (1981, p. 212) calls it) the corpuscular hypothesis restricts those inherent properties to size, shape, number, and motion, and holds that all other qualities and operations are explicable in terms of that restricted set of properties. The orthodox version thus implies a proviso of contact action—that bodies causally interact only locally, by impact, such that unmediated action at a distance is denied. (Although a number of commentators use the terms ‘corpuscular hypothesis’ and ‘mechanism’ interchangeably, distinguishing them has certain benefits. For instance, it permits us to classify Newton as a corpuscularian theorist of some stripe, and to do so without engaging the debate about whether he adhered to the contact action proviso. The definitions given here also agree largely with those employed by Uzgalis in his entry, “John Locke”.) Plenist and atomist versions of the corpuscular hypothesis may be distinguished. Plenist theorists deny the void and assert a plenum of matter, as Descartes does by identifying matter with extension. Such theorists may speak of particles, but their particles are not atoms, being infinitely or at least indefinitely divisible. Atomist theorists, by contrast, accept the void and take the particles or corpuscles comprising compound bodies to be indivisible, or at least probably so. Since Locke's sympathies are clearly with the atomist version, the term ‘corpuscular hypothesis’ shall refer to that version throughout this article, unless indication is given otherwise. Central theses of the Essay are developed in close conjunction with the corpuscular hypothesis—most notably, the distinction between real and nominal essences, which is developed in connection with the distinction between primary and secondary qualities associated with corpuscular theorists, including Locke's mentor, Robert Boyle. Yet since Locke often treats the hypothesis with skepticism, its status and purpose are a source of controversy.

This article examines questions connected with the two salient features noted, and in connection with the first, it also examines Locke's relationship to Newton, a figure instrumental to the changing conceptions of scientific knowledge. Section 2 addresses questions connected to those changing conceptions. What does Locke take science (scientia) or scientific knowledge to be generally, why does he think that scientia in natural philosophy is beyond the reach of human beings, and what characterizes the conception of human knowledge in natural philosophy that he develops? Section 3 addresses the question provoked by Locke's apparently conflicting treatments of the corpuscular hypothesis. Does he accept or defend the corpuscular hypothesis? If not, what is its role in his thought, and what explains its close connection to key theses of the Essay? Since a scholarly debate has arisen about the status of the corpuscular hypothesis for Locke, Section 3 reviews some main positions in that debate. Section 4 considers the relationship between Locke's thought and Newton's. All citations of An Essay Concerning Human Understanding are indicated by ‘E’, followed by the book and section numbers. Page numbers referring to the Nidditch edition are also provided.

2. Locke on knowledge in natural philosophy: scientia and human knowledge

Locke's great epistemological contribution to philosophy is a conception of human knowledge suitable for the experimental science of his day, one that in natural philosophy at least will replace the old, Aristotelian conception. According to the Aristotelian conception, scientific knowledge—scientia—is certain knowledge of necessary truths, which can in principle be expressed in syllogistic form, the conclusion following from self-evident premises. In the domain of natural philosophy, it is certain knowledge of real essences. While he recognizes that the demands of scientia are too stringent for the new, experimental science, Locke does not seriously entertain a radical skepticism, finding “a very manifest difference between dreaming of being in the Fire, and being actually in it.” (E IV.ii.14, pp.537–538) Nor, however, does he fully jettison the concept of scientia. Instead, he develops a conception of human knowledge appropriate to the new science while retaining scientia as an ideal. Thus as he develops his conception of human knowledge in natural philosophy, scientia continually serves as a foil, an ideal knowledge, attainable for human beings in certain domains, but in natural philosophy possible only for nobler spirits, such as angels. This section begins by reviewing the history of the concept of scientia and the factors that, by undermining it, provide the impetus for Locke's conception of human knowledge in natural philosophy. This section also explains what Locke thinks scientia in natural philosophy would amount to, the obstacles that prevent human beings from achieving it, and the lesser human knowledge that must serve us in its place.

2.1 Historical roots of scientia[1]

The conception of genuine scientific knowledge that Locke inherits and in some manner retains, scientia, has its roots in Aristotle, as set out at the beginning of Book I, §2 of the Posterior Analytics. For Aristotle, only necessary truths are the objects of scientific knowledge, and since scientific knowledge requires knowledge of causes, it also requires knowing that the fact is necessary by knowing its necessary relations to its causes. Among Aristotle's manifold causes, the main one in question here is the formal cause—the nature or essence—as Aristotle indicates elsewhere.[2]

We suppose ourselves to possess unqualified scientific knowledge of a thing, as opposed to knowing it in the accidental way in which the sophist knows, when we think that we know the cause on which the fact depends, as the cause of that fact and of no other, and, further, that the fact could not be other than it is….The proper object of unqualified scientific knowledge is something which cannot be other than it is. (Aristotle, Posterior Analytics, I.2)

The knower's epistemic stance toward the necessary truth and its relation to causes is one of certainty, and the certain fact is demonstrable via a syllogism, specifically, one in which the premises are self-evident, requiring no demonstration themselves.

We do know by demonstration. By demonstration I mean a syllogism productive of scientific knowledge, a syllogism….The premisses must be primary and indemonstrable; otherwise they will require demonstration in order to be known, since to have knowledge, if it be not accidental knowledge, of things which are demonstrable, means precisely to have a demonstration of them. The premisses must be the causes of the conclusion, better known than it, and prior to it; its causes, since we possess scientific knowledge of a thing only when we know its cause; prior, in order to be causes; antecedently known, this antecedent knowledge being not our mere understanding of the meaning, but knowledge of the fact as well. (Aristotle, Posterior Analytics, I.2)

The stipulation that the premises of a scientific demonstration must be indemonstrable, that is, self-evident, leads to a prima facie difficulty. This conception of scientific knowledge is intended to encompass not only conceptual propositions, but also propositions about the real natures or essences of substances, which is to say propositions about the world. As with any other demonstration, a demonstration in natural philosophy must have premises that are self-evident, since otherwise a regress would ensue. Yet it is not clear, at least from a contemporary perspective, how the premises can be self-evident, since they must be experience-based.

Now, it would be overstating matters to say that this problem appears only from a contemporary perspective; there was some recognition of it in the ancient and medieval periods.[3] Still, before the advent of the experimental science, the problem was not strongly felt because experience was understood in a different way. For one thing, the notion of an experiment—an artificially constructed, single event or series of events designed to test for a predicted outcome—did not exist. Moreover, among the scholastic Aristotelians, a single, naturally occurring event could not by itself be regarded as revealing of natural processes, because it might be a “monster”—it might be an event that went against nature rather than being produced by and thus revealing of nature.[4] Events as generally experienced, however, were regarded as being revealing of nature, and so they could provide the universal truths needed as premises in the syllogism.[5] How was the gap between events as usually experienced—which still amounts to a limited sample of evidence—and the universal claim derived from them bridged? It is only from the modern and contemporary perspectives that such a gap exists to be bridged. For Aristotle and the medievals, human faculties are so constituted as to be able to apprehend nature, that is, to discern the essences of substances.[6] In short, then, because the internal essences that form the content of natural philosophy are real, and because our faculties are so constituted as to apprehend those real essences, natural philosophy can be a science, a domain in which certain, demonstrative knowledge can be had, despite its dependence upon experience.

The exemplars of scientific knowledge, to be sure, are conceptual disciplines, not only geometry, but also rational theology, the latter being the quintessential science for the medievals.[7] Yet for the most part, natural philosophy stands side by side with conceptual disciplines, even into the modern period. Bacon, though associated with induction, accepts the demonstrative conception of scientific knowledge, and so does Galileo, who understands the role of his experiments accordingly; their purpose is not to test predictions, but to reveal fundamental principles.[8] As empirical methods are refined and more widely applied, the belief that natural philosophy can stand under the umbrella of scientia comes under increasing pressure.[9] Some thinkers resist the pressure, most notably Descartes, who derives his laws of nature by an a priori reflection upon God's nature, and, placing confidence in these rationalist methods, denies that his third law of nature is undermined by conflicting observations of colliding bodies.[10] Yet for the experimentalists themselves, including Locke's mentor, Boyle, observations and experiments are the vehicle to knowledge. It is this approach that puts natural philosophy on the path to Hume's problem of induction, and that most influences Locke.[11]

2.2 Scientia in natural philosophy and obstacles to human attainment

As indicated earlier, scientia serves as the backdrop against which Locke develops the conception of the knowledge that is possible in natural philosophy for human beings. Having retained the ideal of scientia, as exemplified by geometry, yet having also absorbed the import of Boyle's experimental method, Locke is driven to his characteristic pessimism about the kind and extent of knowledge possible for us in natural philosophy. “The meanest, and most obvious Things that come in our way, have dark sides, that the quickest Sight cannot penetrate into” (E IV.iii.22, p. 553). Due to the weakness of our faculties, Locke suspects, “natural Philosophy is not capable of being made a Science.” (E IV.xii.10, p. 645). This section considers Locke's general notion of scientia, what would be required for scientia in natural philosophy, and the obstacles that prevent human beings from attaining scientia in natural philosophy.

2.2.1 Scientia in general

Locke takes knowledge generally to consist in “the perception of the connexion and agreement, or disagreement and repugnancy of any of our Ideas” (E IV.i.1–2, p. 525), and among the three kinds of knowledge that he distinguishes, intuitive, demonstrative, and sensitive, the former two are kinds of certain knowledge. Intuitive and demonstrative knowledge differ in the number of intuitions involved, and consequently, differ in their degree of certainty (E IV.ii.14, pp.537–538). Intuitive knowledge is the most certain because the truth is grasped immediately. There are no intermediate steps, and doubt is impossible because the mind can no more avoid recognizing the truth than the open, functioning eye could avoid seeing light when turned toward the sun (E See IV.ii.1, p. 531). Demonstrative knowledge, though also qualifying as certain, is less so because it involves intermediate steps. We cannot grasp immediately that the three angles of any triangle are equal to two right triangles, and must instead construct the steps of a proof. Upon doing so, and upon grasping the connections between the steps of the proof, we have demonstrative knowledge (E IV.ii.2–3, p.531–532). Intuitive and demonstrative knowledge are forms of scientia, then. Locke defines that as “certain universal Knowledge”(E IV.iii.29, p. 559), and mere “particular matters of fact” (E IV.iii.25, pp. 555–56) do not qualify.

To understand Locke's notion of scientia, we must consider its objects: real essences and the necessary connections that flow from them. Denying the Aristotelian view that a single essence both grounds the properties of a thing, making it what it is, and provides the basis for classifying it, Locke draws a distinction between real and nominal essences.[12] Whereas the nominal essence consists in the set of observable qualities we use to classify a thing (which implies that the nominal essence could vary across time or communities), the real essence (or real or internal constitution, as Locke sometimes writes) is that which makes a thing what it is.

Essence may be taken for the being of any thing, whereby it is what it is. And thus the real internal, but generally in Substances, unknown Constitution of Things, whereon their discoverable Qualities depend, may be called their Essence. This is the proper original signification of the Word, as is evident from the formation of it; Essentia, in its primary notation signifying properly Being. And in this sense it is still used, when we speak of the Essence of particular things, without giving them any Name. (E III.iii.15, p. 417)

Discussion of real essence often focuses upon the real essences of material substances, and in that case we say that the real essence is the causal ground of the substance's perceivable properties. As will be discussed in a subsequent section, many commentators interpret Locke as identifying a material substance's real essence with some subset of its constituent corpuscles' primary qualities, an identification presuming that Locke accepts the corpuscular hypothesis (e.g., Osler 1970, p. 12; Mandelbaum 1964, p.1). According to another interpretation, the real-nominal essence distinction is metaphysical, and thus more fundamental than the primary-secondary quality distinction, which is a physical distinction, belonging as it does to a particular physical theory, the corpuscular hypothesis. Yet leaving that debate aside for the moment, we may note from the first sentence of the above-quoted passage that Locke does not restrict the notion of real essence to substances. This means that we can speak of, say, the real essence of a triangle, understanding it as that which grounds the triangle's qualities, making it what it is.

We have scientific knowledge of something when we know its real essence and, since its qualities flow from that real essence, when we know the necessary connections between the essence and its other qualities. Geometry serves as an exemplar, as it did for so many of Locke's predecessors. In knowing what a triangle is, we cannot conceive things being otherwise than that the sum of its three angles equal the sum of two right angles.

Such knowledge is so certain that we cannot conceive even of God having made things otherwise: “Thus the Idea of a right-line Triangle necessarily carries with it an equality of its Angles to two right ones. Nor can we conceive this Relation, this connexion of these two Ideas, to be possibly mutable, or to depend on any arbitrary Power, which of choice made it thus, or could make it otherwise.” (E IV.iii.29, pp. 559–560)

Scientia is possible in another conceptual domain also: morality. Morality is characterized by discernible necessary connections, and Locke is adamant that we can have the same level of certainty as in geometry.

Where there is no Property, there is no Injustice, is a Proposition as certain as any Demonstration in Euclid: For the Idea of Property, being a right to any thing; and the Idea to which the Name Injustice is given, being the Invasion or Violation of that right; it is evident, that…I can as certainly know this Proposition to be true, as that a Triangle has three Angles equal to two right ones. (E IV.iii.18, pp. 549–50).

2.2.2 Scientia in natural philosophy

What would be required for scientia in natural philosophy? Since scientia generally concerns real essences, and since natural philosophy for Locke concerns material substances and their powers, scientia in natural philosophy would be knowledge of material substances' real essences and their necessary connections to qualities flowing from them.

If we could have scientia in natural philosophy, we could know a substance's qualities without making observations or experiments. To take one of Locke's frequent examples, if we could know gold's real essence, we would then know its qualities, even if not a single sample of gold existed.

Had we such Ideas of Substances, as to know what real Constitutions produce those sensible Qualities we find in them, and how those Qualities flowed from thence, we could, by the specifick Ideas of their real Essences in our own Minds, more certainly find out their Properties, and discover what Qualities they had, or had not, than we can now by our Senses: and to know the Properties of Gold, it would be no more necessary, that Gold should exist, and that we should make Experiments upon it, than it is necessary for the knowing the Properties of a Triangle, that a Triangle should exist in any Matter, the Idea in our Minds would serve for the one, as well as the other. (E IV.vi.11, p. 585)

Which qualities exactly would we be able to deduce? We would be able to deduce a substance's tertiary qualities, that is, its powers to produce certain effects in other substances.[13] If we knew the real essences of opium, and of hemlock, then just as though we were performing a geometric deduction, or just as a locksmith understands why a given key will open one lock but not another, we could deduce that opium produces sleep, that hemlock causes death, and we would understand why each substance produces its effects.

I doubt not but if we could discover the Figure, Size, Texture, and Motion of the minute Constituent parts of any two Bodies, we should know without Trial several of the Operations one upon another, as we do now the Properties of a Square, or a Triangle. Did we know the Mechanical affections of the Particles of Rhubarb, Hemlock, Opium, and a Man, as a Watchmaker does those of a Watch, whereby it performs its Operations, and of a File which by rubbing on them will alter the Figure of any of the Wheels, we should be able to tell before Hand, that Rhubarb will purge, Hemlock kill, and Opium make a Man sleep….The dissolving of Silver in aqua fortis, and Gold in aqua Regia, and not vice versa, would be then, perhaps, no more difficult to know, that it is to a Smith to understand, why the turning of one Key will open a Lock, and not the turning of another. (E IV.iii.25, pp. 555–56) (Cf. Boyle, who had had the same idea, explaining it at length in The Origin of Forms and Qualities, 1666, pp. 16–19.)

Knowing real essences would enable us to deduce tertiary qualities, then, but what about secondary qualities? Here matters are initially less clear, since Locke seems to be saying both that more acute senses would not do away with the secondary quality of sound, but might do away with the secondary quality of color. In one passage, Locke imagines our having very acute senses, including “microscopical eyes”, and he clearly assumes that we would still experience the secondary quality of sound: “If our Sense of Hearing were but 1000 times quicker than it is, how would a perpetual noise distract us” (E II.xxiii.12, pp. 302–303). Yet in a preceding passage, he had suggested that if our faculties were designed for detecting real essences, we would not experience color at all:

Had we Senses acute enough to discern the minute particles of Bodies, and the real Constitution on which their sensible Qualities depend, I doubt not but they would produce quite different Ideas in us; and that which is now the yellow Colour of Gold, would disappear, and instead of it we should see an admirable Texture of parts of a certain Size and Figure. This Microscopes plainly discover to us: for what to our naked Eyes produces a certain Colour, is by thus augmenting the acuteness of our Senses, discovered to be quite a different thing; and the thus altering, as it were, the proportion of the Bulk of the minute parts of a coloured Object to our usual Sight, produces different Ideas, from what it did before….Blood to the naked Eye appears all red; but by a good Microscope, wherein its lesser parts appear, shews only some few Globules of Red, swimming in a pellucid Liquor; and how these red Globules would appear, if Glasses could be found, that yet could magnify them 1000 or 10000 times more, is uncertain. (E II.xxiii.11, pp. 301–302)]

Reflecting upon these examples, however, suggests that he is not, or not always, imagining that microscopical eyes would do away with color altogether; rather, it might in some cases enable us to view tinier particles having different colors than that we perceive for the aggregate object. For in the example he gives, blood as an aggregate body appears uniformly red, but under a microscope, some parts of it appear translucent while only the globules appear red. Color is not eliminated from the experience of seeing blood, once the microscope is used, but is instead seen as being differently distributed. A pixilated painting provides a rough analogy; a shape seen from afar may appear uniformly green, but up close is seen as comprising tiny blue and yellow dots.

2.2.3 Obstacles to human attainment

Scientia in natural philosophy would require knowledge both of real essences and of their necessary connections among qualities, yet neither is possible for human beings, Locke concludes. One obstacle to scientia, then, is that real essences escape us. God has given us sensory capacities that are suitable for finding our way to the “market and exchange” and other practical needs, but are not suitable, as the “microscopical eyes” passage indicates, for detecting the minute parts of bodies.

Another obstacle is that we are almost entirely unable to discern the necessary causal connections among the qualities of substances (and Locke does take those connections to be necessary). Locke does find two instances in which we can discern necessary connections between qualities of bodies: “Some few of the primary Qualities have a necessary dependence, and visible connexion one with another, as Figure necessarily presupposes Extension, receiving or communicating Motion by impulse, supposes Solidity.” (E IV.iii.14, p. 546) Apart from these two exceptions, however, necessary connections escape us. In part, this is due to the first obstacle, our inability to discover real essences, due to the minuteness of particles. It is also due, however, to the remoteness of so many bodies, those which lie “beyond this our Earth and Atmosphere…even beyond the Sun, or remotest Star our Eyes have yet discovered” (E IV.vi.11, p.586–87, and IV.vi.12, p.587). For according to Locke's speculations, all things might be causally interconnected in complex ways, such that we could not know one without knowing all of the others with which it is causally connected.

But while human beings cannot attain scientia in natural philosophy, there are other epistemic agents who can. One is, of course, God, who certainly knows real essences (E III.vi.3, p. 440), and “'tis possible Angels have” ideas of real essences as well (E III.vi.3, p. 440).

It be not to be doubted, that Spirits of a higher rank than those immersed in Flesh, may have as clear Ideas of the radical Constitution of Substances, as we have of a Triangle, and so perceive how all their Properties and Operations flow from thence, but the manner how they come by that Knowledge, exceeds our Conceptions. (E III.xi.22, p. 520)

That Locke finds it natural to speak in the same breath of matter and spirits marks him as belonging to the age of natural philosophy rather than of science. It is because knowledge of necessary connections can be referred to these higher epistemic agents that scientia persists so strongly as an ideal, even while he recognizes the need for a quite different conception of knowledge.

2.3 Human knowledge in natural philosophy (sensitive knowledge)

The conclusion that no intuitive or demonstrative knowledge of substances is possible for us, their real essences and necessary connections being out of reach, places Locke at a crossroads. One path is skepticism, the view that without certainty, no knowledge of substances is possible at all. He rejects that path, denying that hyperbolic doubt could be genuine for either the self (E IV.ix.2, pp. 619–20) or for external objects (E IV.xi.3, p. 631). The other path, the one he follows, involves lowering the bar by admitting a third kind of knowledge, one that lacks certainty: sensitive knowledge.

Sensitive knowledge is knowledge of the “effects [that] come every day within the notice of our Senses”, without an understanding of their causes; “we must be content to be ignorant of” those causes.“(E IV.iii.29, pp. 559–560). Instead of knowing real essences, the causal basis of the properties we perceive, we know only those perceived properties, from which we construct nominal essences. Instead of employing deduction, we are forced to rely upon ”trials“—observations and induction. Instead of knowing the necessary connections holding between a substance's real essence and its other qualities, including its tertiary qualities (which might include, recall, causal connections with substances beyond the remotest star), we know only the co-existences of properties. And from the mere, regular co-existence of properties found in observed cases, Locke observes, we could not know with certainty that the same set will be found co-existing in the next case.

For all the Qualities that are co-existent in any Subject, without this dependence and evident connexion of their Ideas one with another, we cannot know certainly any two to co-exist any farther, than Experience, by our Senses, informs us. Thus though we see the yellow Colour, and upon trial find the Weight, Malleableness, Fusibility, and Fixedness, that are united in a piece of Gold; yet because no one of these Ideas has any evident dependence, or necessary connexion with the other, we cannot certainly know, that where any four of these are, the fifth will be there also, how highly probable soever it may be. (E IV.iii.14, p. 546)

Yet our discoveries about co-existences of properties—though they are merely contingent particulars, or, insofar as they are applied beyond the particular cases we have actually observed, are mere probability—can yet qualify as real knowledge. To do so, our ideas must meet certain conditions. The complex idea that we refer to a substance must comprise all and only those simple ideas that we have found to co-exist in nature. Here Locke is concerned to show that sensitive knowledge deserves its appellation, since it can be distinguished from arbitrary or otherwise poorly grounded claims (e.g., that fluidity has been found to co-exist with brittleness, in a single substance and at a single given temperature.[14]) Sensitive knowledge is far less than scientia, but far more than ungrounded opinion.

Herein therefore is founded the reality of our Knowledge concerning Substances, that all our complex Ideas of them must be such, and such only, as are made up of such simple ones, as have been discovered to co-exist in Nature. And our Ideas being thus true, though not, perhaps, very exact Copies, are yet the Subjects of real (as far as we have any) Knowledge of them. (E IV.iv.12, p. 568)

As for general claims about substances based upon observed particular matters of fact, these too can qualify as real knowledge. Admittedly, it is only probable that, when four of the five properties previously found co-existing together occur again, the fifth will be present too. Yet we can still form an abstract idea of gold, as a substance having all five properties, and call this general claim knowledge, on the grounds that ”whatever have once had an union in Nature, may be united again.“ (E IV.iv.12, p. 568)

Has contemporary science enabled us to go beyond sensitive knowledge—have discoveries about compounds, elements, and subatomic particles provided us with knowledge of real essences? Much of the force of this question derives, to paraphrase Nicholas Jolley, from the fact that many of those discoveries about matter's structure were not conceived empirically, but only confirmed empirically; they were initially conceived as possibilities by employing the hypothetico-deductive model, and the predictions deduced from the models were then compared to empirical data (Jolley 2002, p. 69). Yet as Jolley also points out, these commentators may have missed the full import of Locke's geometric model; in a passage quoted earlier, Locke tells us explicitly that if we knew the real essence of gold, we could deduce its qualities even if gold did not exist.[15] So while the predictions of any model developed via the hypothetico-deductive model must survive the test of observations, observations in Locke's scientia are wholly unnecessary. To put the point another way, Locke takes natural philosophy to be an empirical domain only for human beings, with their impoverished faculties. For nobler spirits, it would more resemble geometry.

3. Tension in Locke's thought and a consequent debate

The last few decades have seen a lively debate about the role of the corpuscular hypothesis in Locke's Essay. This section examines the sources of that debate and reviews some of the main positions figuring in it.

3.1 Tension in Locke's thought

As we have seen, Locke develops some central theses of his Essay in connection with the corpuscular hypothesis. In his theory of ideas, corpuscles provide at least a structural basis for simple ideas, and depending upon one's interpretation, there may be a causal relationship as well. Further, and of particular interest here, Locke often appears to identify a material substance's real essence with the set or some subset of its component particles' primary qualities. In the following well-known passage, for instance, he points to the primary qualities of a body's parts—their bulk or solidity, motion, and shape—as the causal ground of the qualities we perceive.

The particular Bulk, Number, Figure, and Motion of the parts of Fire, or Snow, are really in them whether any ones Senses perceive them or no: and therefore they may be called real Qualities, because they really exist in those Bodies. But Light, Heat, Whiteness, or Coldness, are no more really in them, than Sickness or Pain is in Manna. Take away the Sensation of them; let not the Eyes see Light, or Colours, nor the Ears hear Sounds; let the Palate not Taste, nor the Nose Smell, and all Colours, Tastes, Odors, and Sounds, as they are such particular Ideas, vanish and cease, and are reduced to their Causes, i.e. Bulk, Figure, and Motion of Parts. (E II.viii.17, pp. 137–138)

He similarly seems to identify the real essence of bodies with primary qualities when he suggests, just prior to the ”microscopical eyes“ passage, that instead of seeing colors (or instead of seeing them as we currently do), we could discover bodies' internal constitutions, if only we knew the ”texture and motion of the minute Parts of corporeal things“ (E II.xxiii.12, pp. 302–303). And a commitment to the corpuscular hypothesis is again suggested when he despairs of understanding the production of secondary qualities: even if ”we could discover the size, figure, or motion of those invisible parts, which immediately produce them [secondary qualities}, we still cannot discover any “undoubted Rules” concerning their production or connection, nor “conceive how any size, figure, or motion of any Particles, can possibly produce in us the Idea of any Colour, Taste, or Sound” (E IV.iii.13, p. 545). Here he appears to despair of understanding how secondary qualities are produced by primary ones; he appears to take the corpuscular hypothesis' reductionist claim to be true, but he despairs of our understanding how the reduction works.

His discussion of tertiary qualities is similar. If we knew the “Figure, Size, Texture, and Motion of the minute Constituent parts of any two Bodies”, we would then be able to derive tertiary qualities; we would be able to deduce that opium causes sleep, and we would understand why (E IV.iii.25, pp. 555–56; see also E IV.iii.13, p. 545). In all of these passages, then, and in many similar ones, Locke at least appears to accept at least some components of the corpuscular hypothesis—that material bodies are compounded from minute particles, and that certain observable qualities are reducible to the particles' primary qualities of size, shape, and motion. This tendency to speak as though the corpuscular hypothesis is true, either in whole or in part, has been termed Locke's “dogmatic” side (Downing 2007).

In apparent tension with this so-called dogmatic side is what has been termed his “agnostic” or “skeptical” side. The following features of his discussion seem to suggest that he has reasons either for remaining agnostic about whether the corpuscular hypothesis is true, or more seriously, for believing that it is wholly unable to explain the phenomena it purports to explain and therefore cannot be true.

First, he refers to the corpuscular hypothesis as such, a hypothesis, one that falls well short of providing us with scientific knowledge; and further, he remarks that it is not his aim to adjudicate among competing hypotheses.

I have here instanced in the corpuscularian Hypothesis, as that which is thought to go farthest in an intelligible Explication of the Qualities of Bodies; and I fear the Weakness of humane Understanding is scarce able to substitute another, which will afford us a fuller and clearer discovery of the necessary Connexion, and Co-existence, of the Powers, which are to be observed united in several sorts of them. This at least is certain, that which ever Hypothesis be clearest and truest, (for that it is not my business to determine,) our Knowledge concerning corporeal Substances, will be very little advanced by any of them, till we are made see, what Qualities and Powers of Bodies have a necessary Connexion or Repugnancy one with another; which in the present State of Philosophy, I think, we know but to a very small degree. (E IV.iii.16, pp. 547–548)

The hypothetical status of all physical theories is underscored also in Some Thoughts Concerning Education: “The systems of natural philosophy…are to be read, more to know the hypotheses…than with hopes to gain thereby a comprehensive, scientifical, and satisfactory knowledge of the works of nature.”(Locke, quoted in Rogers 1982, p. 230.)

Second, if Locke indeed identifies material bodies' real essences with the primary qualities of their constituent corpuscles, then that view of real essence together with his pessimism about the possibility of ever discovering real essences imply pessimism about the corpuscular hypothesis—specifically, about the claims that bodies are made of corpuscles and that their observable qualities are reducible to the corpuscles' qualities. In the same passages in which Locke seems to accept or assume these central tenets of the corpuscular hypothesis—that observable bodies are made up of corpuscles and that those corpuscles have a restricted set of inherent properties—he simultaneously appears very skeptical about the hypothesis' promise to reduce observable properties such as color and taste to that restricted set of primary properties.

Third, Locke arguably takes the corpuscular hypothesis to have limitations or shortcomings so serious that they amount to fatal flaws, an interpretation that Margaret Wilson (1979) was perhaps the first to defend. Wilson develops her line of argument mainly in connection with difficulties Locke raises about the corpuscular hypothesis' purported ability to explain sensation and more generally, the relation between thought and matter,[16] but some other phenomena are troublesome as well. Locke appears to consider such phenomena to be so obscure that we can attempt to understand them only by attributing them to God's direct action.

The coherence and continuity of the parts of Matter; the production of Sensation in us of Colours and Sounds, etc. by impulse and motion; nay, the original Rules and Communication of Motion being such, wherein we can discover no natural connexion with any Ideas we have, we cannot but ascribe them to the arbitrary Will and good Pleasure of the Wise Architect. (E IV.iii.29, pp. 559–560)

Elsewhere, Locke will use the term ‘superaddition’ to refer to God's role. Roughly speaking, superadded properties are those properties specifically added by God.

3.2 Limitations of the corpuscular hypothesis

This section examines the phenomena that Locke seems to consider too obscure for the corpuscular hypothesis to illuminate. These are the three phenomena mentioned in the above-quoted passage, the production of sensation, the communication of motion, cohesion,[17] and a fourth, gravity, which Locke discusses directly only outside the Essay. A subsequent section reviews some of the main positions that have been taken in the debate about the status of the corpuscular hypothesis for Locke, and that same section considers various interpretations of superaddition, since any interpretation of that concept is logically tied to one's view about Locke's stance toward the corpuscular hypothesis.

3.2.1 Sensation

As we saw in passages discussed earlier, in connection with the impossibility of scientia, Locke finds the production of sensation to be utterly obscure. One side of the difficulty is, of course, the nature of the mind. In all probability, it is immaterial. However, Locke allows, it is possible that God has superadded the power of thought directly to matter. The other side of the difficulty concerns the nature of secondary qualities as powers to produce sensations. Much of the persuasiveness of the corpuscular hypothesis lay in its reductive promise. Secondary qualities in particular, such as colors and sounds, but also ideas of macrolevel primary qualities, including visual sensations of shapes and sizes, and tertiary qualities were to be reduced to the primary qualities of corpuscles, interacting with one another and/or with perceptual systems.

One part of the corpuscular hypothesis' purported explanation is conceivable, namely, the interactions among the primary qualities of bodies, which are supposed to be part of the causal basis of our sensations:

That the size, figure, and motion of one Body should cause a change in the size, figure, and motion of another Body, is not beyond our Conception; the separation of the Parts of one Body, upon the intrusion of another; and the change from rest to motion, upon impulse; these, and the like, seem to us to have some connexion one with another. (E IV.iii.13, p. 545.)

Indeed, we are able to discern necessary connections in two instances, as noted earlier. (One case involves only primary qualities— “Figure necessarily presupposes Extension” (E IV.iii.14, p. 546)—and the other involves tertiary and primary qualities—“receiving or communicating Motion by impulse, supposes Solidity” (E IV.iii.14, p. 546).) If we knew more about the primary qualities of bodies, we might multiply such instances: “And if we knew these primary Qualities of Bodies…we might be able to know a great deal more of these Operations of them one upon another.” That is, if we knew real essences, we could derive more necessary connections, knowing, for instance, the causal relation between opium and sleep, and as certainly as we now know that impulse requires solidity.

Yet knowing real essences would not give us any genuine knowledge of how sensations are produced by primary qualities. While corpuscular theorists such as Galileo (The Assayer) sketched a reductive account of our sensations of taste in terms of particles striking our tongues, Locke suggests that any attempt to discover the process's details will be foiled. For as far as we can imagine, a body, by striking other bodies, is able to produce “nothing but Motion” [18] (E IV.iii.6, pp. 540–541), and motion may itself be hopelessly obscure, as indicated below. As for shape and size, we can no more imagine how they could figure in the production of sensations than we can imagine how motion could.

We are so far from knowing what figure, size, or motion of parts produce a yellow Colour, a sweet Taste, or a sharp Sound, that we can by no means conceive how any size, figure, or motion of any Particles, can possibly produce in us the Idea of any Colour, Taste, or Sound whatsoever; there is no conceivable connexion betwixt the one and the other. (E IV.iii.13, p. 545.)

Although Locke mentions only secondary qualities here, his point presumably applies to all sensations, including our sensations of macrolevel primary qualities, such as the shape and size of a snowball or lump of gold. For again, genuine knowledge is knowledge of necessary connections, with the conceptual relations in geometry being the model, and it does not seem possible to discover such connections between any sensation and the sizes, shapes, and textures that are supposed to cause them. Despite the resemblance of macrolevel primary qualities to microlevel ones, both being of the same type,[19] the idea of a quality is nonetheless a very different thing than the quality itself.

Locke finds that our only way of understanding the production of sensation is to attribute the process to God. If we try to understand how motion could produce a color, sound, or taste, “we are fain to quit our Reason, go beyond our Ideas, and attribute it wholly to the good pleasure of our Maker.” (E IV.iii.6, p. 540–541; see also IV.iii.28, p. 559.)

3.2.2 Gravity

With the publication of Newton's Principia, implying as it did the possibility of unmediated action at a distance, gravity becomes the most nettlesome phenomenon for the orthodox version of the corpuscular hypothesis, which includes a proviso of contact action. Locke is initially sympathetic to the provise, writing in the first three editions of his Essay, “How bodies operate one upon another…is manifestly by impulse and nothing else. It being impossible to conceive that body should operate on what it does not touch.” (E II.viii.11, editions 1–3) Yet for the fourth edition, he replaces that claim about how bodies do operate with one about how we can conceive of them operating: “How Bodies produce ideas in us is manifestly by impulse, [this being] the only way which we can conceive Bodies [to] operate.”(E II.viii.11, edition 4). He also omits a clause appearing in II.viii.12 of previous editions that denied unmediated action at a distance.[20] These subtle emendations reflect a dramatic shift, one expressed directly in his correspondence with Stillingfleet.

The gravitation of matter towards matter, by ways inconceivable to me, is not only a demonstration that God can, if he pleases, put into bodies powers and ways of operation, above what can be derived from our idea of body, or can be explained by what we know of matter, but also an unquestionable and every where visible instance, that he has done so. (Second Reply to the Bishop of Worcester, 1699, The Works of John Locke, Vol. IV, p. 467)

The phenomenon of gravity—as explained by “Mr. Newton's incomparable book”(ibid.)—has apparently led Locke to abandon the contact action proviso and to attribute to matter the power of acting distantly, even though the process by which such interactions could occur is so obscure that he is driven to invoking superaddition. (Or at least, so goes the prevailing interpretation; not all commentators agree, as indicated in a subsequent section.)

3.2.3 Motion

Locke takes the notion of impulse, in which bodies communicate motion to one another by surface impact, to be, along with extension and cohesion, fundamental to our concept of body.[21] Indeed, regardless of how motion may in fact be communicated, impulse is the only means by which we can conceive of its being communicated, a view Locke maintains, as we saw, despite his changing thoughts about gravity. Impulse is also fundamental to the corpuscular hypothesis' explanation of phenomena, being either the exclusive means of interaction among bodies, as adherents of the contact action proviso hold, or the means of at least many interactions. Yet how exactly does a moving body communicate motion to a resting one simply by impacting it? When we attempt to discover the precise nature of the process, Locke suggests, we find that it is just as mysterious as the process by which the mind moves the body.

Another Idea we have of Body, is the power of communication of Motion by impulse; and of our Souls, the power of exciting of Motion by Thought…. But if here again we enquire how this is done, we are equally in the dark. For in the communication of Motion by impulse, wherein as much Motion is lost to one Body, as is got to the other, which is the ordinariest case, we can have no other conception, but of the passing of Motion out of one Body into another; which, I think, is as obscure and unconceivable, as how our Minds move or stop our Bodies by Thought….The increase of Motion by impulse, which is observed or believed sometimes to happen, is yet harder to be understood. We have by daily experience clear evidence of Motion produced both by impulse, and by thought; but the manner how, hardly comes within our comprehension; we are equally at a loss in both. (E II.xxiii.28, p. 311)

Since the corpuscular hypothesis holds impulse to be the primary if not the sole means by which bodies causally interact, then any phenomenon that the corpuscular hypothesis purports to explain by impulse is obscure if impulse itself is obscure. All of the hypothesis' reductions of observable primary qualities, of secondary qualities, and of tertiary qualities, would inherit impulse's obscurity; thus Locke appears to be suggesting here that the corpuscular hypothesis cannot fulfill its promise of explaining and reducing those properties and powers.

3.2.4 Cohesion

Since the claim that observable bodies are made up of particles is central to the corpuscular hypothesis, an immediate question for its proponents asks how the particles cohere into compound bodies. Plenists typically have some resources for answering that question. Descartes, for instance, though he speaks in terms of particles, understands an individual body as an area of extension moving as one with respect to surrounding areas. It is not possible for any particle to move away into empty space, there being no such thing as empty space once matter is identified with extension; with every bit of matter pressed from all sides by other matter, there is no problem about cohesion. In the same vein, Malebranche can invoke the pressure of the air to explain the coherence of bodies, and then invoke the pressure of an aether to explain the coherence of air particles. Locke objects that this explanation fails because it leaves us with the question of what causes the particles of the aether to cohere (E II.xxiii.23, p, 308). The objection reveals Locke's atomist sympathies, drawing its power from the presumption that there is such a thing as empty space into which the aether particles could move. The problem for atomist versions of the corpuscular hypothesis is that the restricted set of properties that they allow the particles —size, shape, and motion—provides no obvious resources for explaining how the particles cohere with one another to form compound bodies. In various forms, the problem about cohesion has dogged atomists since ancient times.

The problem arises in two forms, which, to borrow James Hill's terminology (Hill 2004), may be called the limited and the foundational problems. The limited problem, arising for those who take corpuscles to be genuine atoms, that is, to be indivisible, is the problem of explaining how those indivisible corpuscles cohere with one another. This is the problem one finds in Newton's writings. Although Rule 3 of the Principia allows the possibility that the least parts of matter could turn out to be divisible, his atomist sympathies are evident throughout his writings. He speculates in Query 31 that in all probability, bodies are made up of hard particles that only God could divide, and in the body of the Opticks (Book II, Part III, Proposition VII) he suggests that more powerful microscopes might permit us to see the larger particles. In answer to the problem of how those naturally indivisible particles cohere, he rejects the ancient solution of hooked particles as begging the question, proposing instead some short-range forces modeled on the gravitational force (Query 31). Newton's speculations about such forces are driven by an absence of any resolution to the problem about cohesion within the corpuscular theory itself.

The foundational problem pushes the question about cohesion into the corpuscles themselves. The problem was raised by Joseph Glanvill: “If it be pretended…that the parts of solid bodies are held together by hooks, and angulous involutions; I say, this comes not home: For the coherence of the parts of these hooks…will be of as difficult a conception, as the former.”(Glanvill, The Vanity of Dogmatizing, p. 18, quoted in Hill 2004, p. 616) Without any grounds for asserting that the divisibility of matter bottoms out in indivisible corpuscles, then, the question arises of how the parts of a corpuscle could cohere, how the parts of those parts could cohere, and so on, ending in the question of how extended bodies are possible at all.

Whatever his ultimate view of the corpuscular hypothesis, Locke inevitably faces the atomist's problem about cohesion, accepting as he does void space (see II.xiii.11, 12–14, 21–23), and holding that our ideas of body depend fundamentally upon cohesion. One of the ideas “proper and peculiar” to body, he writes, is “the cohesion of solid, and consequently separable parts” (E II.xxiii.17, p. 306), and the extension of body, as opposed to the extension of space, is “nothing, but the cohesion or continuity of solid, separable, moveable Parts”(E II.iv.5, p. 126). Yet we have no understanding of cohesion, and so our idea of body does not rest on any genuine understanding of it. In trying to understand how bodies are extended, we are as much in the dark as when we try to understand how the soul thinks.

'Tis as easie for him to have a clear Idea, how the Soul thinks, as how Body is extended. For since Body is no farther, nor otherwise extended, than by the union and cohesion of its solid parts, we shall very ill comprehend the extension of Body, without understanding wherein consists the union and cohesion of its parts; which seems to me as incomprehensible, as the manner of Thinking, and how it is performed. (E II.xxiii.24, p. 309)

One scholarly debate about cohesion concerns the question of whether Locke acknowledged only the limited problem, as one would expect from commentators who read him as accepting atomism (e.g. Mandelbaum 1964, p. 1), or whether he looked further, to the more serious, foundational problem, as Hill argues (2004). A related controversy, to be discussed in the next section, concerns the questions of whether Locke concludes that the corpuscular hypothesis simply cannot resolve the problem (e.g., Hill 2004; Downing 2007, p. 408), or whether he remains agnostic on the issue (e.g., McCann in Chappell 1998, p. 244).

The four problematic phenomena winnow away the components of the corpuscular hypothesis, as defined at the outset of this article. The problem about sensation threatens the corpuscular hypothesis' promise of reducing secondary, tertiary, and macrolevel primary qualities to microlevel primary qualities. Newton's results about gravitational phenomena cast grave doubts upon the contact action proviso, and according to some commentators, those results lead Locke to abandon the belief that impulse is the only means of causal interaction. A corpuscular theorist might hope to preserve some part of the theory by insisting that impulse is still the means by which most other causal interactions are effected; but this runs up against the problem about impulse, in that the process by which motion is communicated seems utterly obscure. Finally, even the core claim that observable bodies are composed of tiny corpuscles is threatened by the problem of cohesion. This last problem threatens to be the most serious of four phenomena, for as James Hill has pointed out (2004, p. 628), the problems about gravity, sensation, and motion arise subsequent to our having conceived of body, whereas the problem about cohesion may thwart our very ability to conceive clearly of body.

3.3 Main positions in the debate

This section considers some main responses to the tension between Locke's seeming acceptance of the corpuscular hypothesis, most notable in his apparent identification of a material substance's real essence with the size, shape, and texture of its insensible parts, and his pessimism about the hypothesis' explanatory power, most notable in his remarks about the four phenomena discussed above.

One approach to the tension is to understand it as a genuine inconsistency. Margaret Wilson has defended such an interpretation, though in the 1979 paper that launched the debate, her intent is to show how acutely Locke understood the explanatory limitations of “Boylean mechanism”. Specifically, Wilson argues, the inconsistency reveals Locke's recognition “that some presumed properties of matter cannot be conceived as ”natural“ consequences of Boylean primary qualities” (Wilson 1979, p. 197). Thus our ignorance about bodies has more profound causes than our ignorance about the primary qualities of a body's constituent corpuscles. In accordance with her view that Locke's agnostic tendencies (just like his dogmatic ones) are genuine, Wilson interprets Locke's concept of superaddition robustly, as a sort of divine action that goes beyond the corpuscular hypothesis. According to this “non-essentialist” or “divine annexation” reading, Locke understands superadded properties as properties that God has annexed to matter by fiat, and that bear no intrinsic connection to matter's real essence. This reading implies a distinction in etiology for superadded qualities. Whereas the other qualities of matter are either given initially, as those qualities constituting the real essence, or else flow from the real essence, superadded qualities are added on after the fact, such that the substance could have been complete without them.

Some other interpretations absolve Locke of inconsistency, either by emphasizing his so-called dogmatic side while downplaying or reinterpreting his agnostic tendencies, or emphasizing his agnosticism while downplaying his dogmatism. One line of interpretation, then, reads Locke as in some manner accepting the corpuscular hypothesis (Mandelbaum 1964, chapter 1; Osler 1970, p. 12; Ayers 1975; McCann 1994, §1 and p. 85; McCann 2002, pp. 354–355). According to weaker versions of this reading, Locke's project is the naturalistic one of pursuing the philosophical implications of the best available scientific theory, and developing an epistemological basis for it. McCann, for example, reads Locke as defending the atomist version of the corpuscular philosophy over its Cartesian competitor by providing an epistemology for it. While Descartes had provided an epistemology for his plenist version, there was nothing comparable for the atomist version associated with Gassendi and Boyle until Locke supplied it (McCann 2002, pp. 354–355). According to Ayers' stronger interpretation, Locke accepts “pure mechanism”, that is, the orthodox version of the corpuscular hypothesis, which includes the contact action proviso. According to this view, all of matter's qualities flow from its real essence (Ayers 1981).

Since this line of interpretation seeks to downplay Locke's agnostic tendencies, one challenge is to account for Locke's pessimism about the possibility of our knowing real essences. Mandelbaum meets the challenge by confining Locke's pessimism to the real essences of particular material substances; we are able to know “the general properties possessed by all material substances”, and are ignorant only of the “particular sizes, shapes, number, or motions of the particles which go to make up any specific object”(Mandelbaum 1964, p. 54). A related challenge is to account for Locke's appeals to superaddition, since prima facie, Locke's reason for invoking God is that he thinks the corpuscular hypothesis has no resources for explaining the four problematic phenomena. Ayers responds by rejecting Wilson's divine annexation interpretation of superaddition in favor of a deflationary one. According to Ayers' “divine architect” interpretation, Locke makes no distinction in etiology by calling a property superadded; he means only that God selected the property with particular care when first creating matter. To diffuse the effect of Locke's remarks to Stillingfleet, in which Locke appears to embrace action at a distance, Ayers points to Locke's late manuscript, “The Elements of Natural Philosophy”, interpreting certain passages as referring gravitational effects to an undetectable medium (Ayers 1981, pp. 212–214). This move has been challenged by Stuart. Claiming that the manuscript was probably written for the education of a child, Stuart denies that it could trump Locke's remarks to Stillingfleet (see Stuart 1998, pp. 378–379).

Another way to absolve Locke of inconsistency is to emphasize his agnosticism or skepticism, while downplaying or reinterpreting passages that appear to commit him to the corpuscular hypothesis. Interpretations in this vein tend to emphasize Locke's pessimism about our ability to know real essences, to discern necessary connections, and consequently, to have scientia in natural philosophy. Commentators pursuing this line include Jolley and Downing.

The central challenge facing such interpretations is to account for the passages in which Locke speaks as though he accepts the corpuscular hypothesis, most notably those in which he appears to identify the real essences of material substances with the corpuscles' primary qualities. Jolley (2002) accounts for Locke's so-called dogmatic tendency in strategic terms. Locke's Essay targets both Aristotelians and Cartesians, and though agnosticism, which targets the Cartesians, is ultimately the dominant tendency in his thought, Locke emphasizes the explanatory power of the corpuscular hypothesis whenever he has the Aristotelians in his sights. Downing (1998, 2007) interprets Locke's Essay as developing metaphysical distinctions that constrain physical theory, and then downplays his dogmatic side by taking the corpuscular hypothesis to be truly only a hypothesis for him, and denying that real essence can be identified with primary qualities. Properly understood, Downing argues, the distinction between real and nominal essence is a metaphysical distinction. It is thus more fundamental than the distinction between primary and secondary qualities, which belongs to a particular physical theory, the corpuscular hypothesis. To be worth its salt, a physical theory must meet the metaphysical constraint provided by the real-nominal essence distinction. That is, the physical theory must provide some way of making sense of the notion that material bodies have an internal constitution that is inaccessible to us but that produces those qualities that are accessible. Locke explains his metaphysical distinction using one physical theory, the corpuscular hypothesis, as an illustration, and he often appears to accept or even defend that hypothesis. Still, this is only an appearance, an appearance due to a certain advantage that the corpuscular hypothesis has over other physical hypotheses: it is the theory best suited to our sensory capacities and understanding. Despite its unique status, Locke sees it as only a hypothesis, one crippled by the explanatory limitations evidenced by the problematic phenomena discussed earlier. For Downing, then, Locke's dogmatic tendencies disappear, leaving only his agnostic side.

4. Locke and Newton

The works of Locke and Newton manifest a noticeable intellectual affinity, and their endeavors have been seen as largely complementary, Locke supplying a detailed epistemology for Newton's natural philosophy. Locke expresses that intention when he describes himself as an under laborer for thinkers such as Newton. Although their thought diverges on certain matters, and though their main works were produced independently of one another—Locke's Essay, published subsequently, was essentially complete by the time he read the Principia—influences may be also detected, and in both directions. Opportunity for intellectual influence increased once the two met (probably in 1689, though the exact date is not known; Westfall 1980, 488; Rogers 1982, p. 219). They established a friendship, exchanging views on a wide variety of subjects, not least certain unorthodox theological convictions (see Westfall 1980, 490–91).

4.1 Epistemology

The most obvious example of Locke's influence upon Newton appears in a draft passage, probably penned shortly after the Prinicipia's second edition; the passage has a Lockean tone, and Newton denies that any ideas are innate.[22] But if the tone and language of that late passage is drawn from Locke, Newton's overall epistemological approach bore similarities to Locke's long before the two men met. Like Locke, Newton holds that, revelation apart, we must gather what knowledge we can from experiments. And like Locke he holds that real essences elude us. In the early manuscript, De Gravitatione, Newton denies knowing the “essential and metaphysical constitution” of matter (Newton, 2004, p. 27), and he retains the position in much later texts, including the 1713 General Scholium “We certainly do not know what is the substance of any thing. We see only the shapes and colors of bodies, we hear only their sounds, we touch only their external surfaces….But there is no direct sense and there are no indirect reflected actions by which we know innermost substances”(Principia, 942).

To the extent that Locke and Newton are sympathetic to atomism, both face the epistemological problem that has come to be known as “transduction”. One of the difficulties associated with atomist versions of the corpuscular hypothesis is that of justifying claims about the corpuscles that are alleged to comprise observable bodies, since the corpuscles are commonly[23] taken to be unobservable even in principle.[24] Newton responds to the problem with his analogy of nature, which rests upon a presumption that nature is uniform. His response is codified in Rule 3 together with the subsequent, explanatory remarks:

The qualities of bodies that cannot be intended and remitted [i.e., qualities that cannot be increased and diminished] and that belong to all bodies on which experiments can be made should be taken as qualities of all bodies universally. For the qualities of bodies can be known only through experiments; and therefore qualities that square with experiments universally are regarded as universal qualities; and qualities that cannot be diminished cannot be taken away from bodies. Certainly idle fancies ought not to be fabricated recklessly against the evidence of experiments, nor should we depart from the analogy of nature, since nature is always simple and consonant with itself. (Principia, Book 3, p. 795)

Yet in keeping with his epistemological modesty, Newton allows the possibility, mentioned earlier, that the atomist presumption may turn out to be incorrect: “If it were established by even a single experiment that in the breaking of a hard and solid body, any undivided particle underwent division, we should conclude by the force of this third rule not only that divided parts are separable but also that undivided parts can be divided indefinitely.” (Principia, Book 3, p. 795) For Locke, the seriousness of the problem of transduction depends largely upon the debate about the status of the corpuscular hypothesis; if he takes an agnostic or skeptical stance toward it, then he owes no solution to the problem.

But though Locke and Newton may be said to have shared an epistemic modesty, the nature and extent of that agreement is unclear insofar as controversy attends Locke's reaction to Newton's Principia. What precisely did Locke take the Principia to have achieved, in epistemological terms?

To pose the question in the starkest possible terms, did Locke categorize Newton's epistemological achievement in natural philosophy as a contribution to sensitive knowledge alone? That is, did he see Newton's “mighty Designs in advancing the Sciences” (Essay, Epistle to the Reader, pp. 9–10) as confined to providing a firm basis for natural philosophy through his experimental method? Or might he have taken Newton's mathematical methods as offering the demonstrations needed to push natural philosophy into the domain of demonstrative knowledge and hence scientia? That is, much as the Principia led Locke to retreat from the contact action proviso, might it also have tempted him to retreat from his belief that natural philosophy cannot be made a science?

Commentators interpreting Locke as seeing Newton's contribution in terms of sensitive knowledge alone include Yolton (1969); Woolhouse (1994); and Downing (1997, see especially pp. 292–93). Kenneth Winkler (2008), however, reads Locke's correspondence with Stillingfleet and other writings from the 1690's as a defense of Newton's mathematical physics, and one that draws from his mathematical demonstrations a greater optimism about the possibility of certain knowledge in natural philosophy. This move has been contested by Domski (2012), who argues that Locke advocated Newton's mathematical methods only in connection with astronomy, whose objects are unavailable for experimentation; and that Locke maintained his emphasis upon natural historical methods for questions about terrestrial bodies.

4.2 Ontology and philosophy of nature

Locke and Newton accept similar ontologies. Although both (quite independently of one another) are characteristically circumspect about the nature of mind, they may be considered substance dualists. Newton, though indicating in the early manuscript De Gravitatione that he does not pretend to know the substantial foundation of minds, consistently presents God and minds as immaterial, lacking certain characteristics of body, namely, hardness, impenetrability and resistance. Similarly, Locke, though he explicitly allows the possibility of thinking matter in his Essay and discusses it at length with Stillingfleet, emphasizes that in all probability, the soul is immaterial (E IV.iii.6, pp. 540–541). But while Newton takes even immaterial spirits to be spatially extended, such that a mind can co-occupy place with a body and God with all bodies, it is not clear that Locke agrees. Agreement might be claimed on the basis of the several passages in the Essay. At E II xxvii.2 (p. 329 ), he locates spirits spatially: “Finite spirits having had each its determinate time and place of beginning to exist, the relation to that time and place will always determine to each of them its identity, as long as it exists.” And as the passage continues, he seems to echo Newton's suggestion that substances of different kinds might share place; although substances of the same kind exclude one another out of the same place, he writes, “these three sorts of substances, as we term them, do not exclude one another out of the same place”. Yet a Cartesian interpretation does not seem impossible here, especially in light of remarks elsewhere; in discussing the possibility of thinking matter, the immaterial soul against which he contrasts it is unextended.[25]

As for the concept of body, their views are again similar, without being identical. Both confront the problem of a substratum or substance in general, one that Locke notoriously finds vexing.[26] Newton dispenses with the problem more easily. In De Gravitatione, he eliminates the unintelligible notion of prime matter by associating perceived properties with determined regions of extension.[27] Both also attack Descartes' identification of matter with extension, instead defending the void and showing strong sympathy for atomism. As indicated just above, Newton does not confine the quality of extension to body, whereas Locke may do so. Further, Newton's list of the qualities of body includes not only extension, hardness, impenetrability, and mobility, but also the vis inertiae—the inherent force or power of resisting (Principia, Definition 3), which some commentators take Newton to identify with mass.[28] Locke's concept, however, emphasizes extension, mobility, and solidity, without mentioning mass. Still, since Locke explicates solidity in terms of impenetrability, which arises from resistance (E II.iv.1) there may be grounds for holding that his concept does after all include mass, a position defended by Woolhouse (2005).[29] At the least, Locke's discussion of action by impulse presupposes mass, as Stein has noted. And yet presupposing the concept may not amount to including it in the concept of body, since as Stein points out, mass for Locke “cannot be construed to correspond to a simple idea, but can only be understood as a power ‘mediately perceivable.’” (Stein 1990, p. 36)

There is also the question about gravity's relation to body, for both thinkers a persistent source of controversy. (For an overview of the debate concerning Newton, see Kochiras (2011, Section 5).) One point of clarity and consensus is that neither Locke nor Newton considers the power of gravitational attraction to be essential to matter. Newton consistently denies that it is essential (for instance in his explanatory remarks following Rule 3 of the Principia) and Locke refers to it only as power superadded or endowed by God. Locke's concept of superaddition is controversial of course, as noted earlier, but he does appear to embrace action at a distance in his letter to Stillingfleet. Newton's writings contain no such dramatic pronouncements. So while a few commentators interpret Newton as accepting action at a distance, grounded either in a superadded property (Henry 1994) or in a relational quality of matter (Schliesser 2011), his more favorable remarks are at best indirect. Most commentators interpret him as having very grave misgivings about distant action, if not outright rejecting the possibility. If that interpretation is correct, then Locke was right to credit his change of heart about gravity to “Mr. Newton's incomparable book”, rather than to Newton himself.

Bibliography

Primary Literature

  • Locke, John, 1975 [1700], An Essay Concerning Human Understanding, P. H. Nidditch (ed.), based on the fourth edition, New York: Oxford University Press.
  • Locke, John, 1824, The Works of John Locke, in nine volumes, 12th edition, London: C. and J. Rivington.
  • Newton, I., 1999 [1726], The Principia: Mathematical Principles of Natural Philosophy, trans. I. Bernard Cohen and Anne Whitman, Berkeley: University of California Press.
  • Newton, I., 2004, Newton: Philosophical Writings, ed. Andrew Janiak, Cambridge: Cambridge University Press, 2004.

Secondary Literature

  • Anstey, Peter, 2011, John Locke and Natural Philosophy, Oxford: Oxford University Press.
  • Atherton, M., 1991, “Corpuscles, Mechanism, and Essentialism in Berkeley and Locke,” Journal of the History of Philosophy, 29 (1): 47–67.
  • Ayers, M.R., 1975, “The Ideas of Power and Substance in Locke's Philosophy,” Philosophical Quarterly, 25 (98): 1–27.
  • –––, 1981, “Mechanism, Superaddition, and the Proof of God's Existence in Locke's Essay,” The Philosophical Review, 90 (2): 210–251.
  • Boyle, Robert. 1666. The Origine of Formes and Qualities (According to the Corpuscular Philosophy), Oxford: H. Hall.
  • Clarke, D.M., 1992, “Descartes' philosophy of science,” in J. Cottingham (ed.), The Cambridge Companion to Descartes, Cambridge: Cambridge University Press, 258–285.
  • Cohen, I.B., 2002, “Newton's concepts of force and mass, with notes on the Laws of Motion”, in I. Bernard Cohen and George E. Smith (eds.), The Cambridge Companion to Newton, Cambridge: Cambridge University Press, 2002, pp. 57–84.
  • Curley, E.M., 1972, “Locke, Boyle, and the Distinction between Primary and Secondary Qualities,” The Philosophical Review, 81 (4): 438–464.
  • Dear, P., 1995, Discipline and Experience: The Mathematical Way in the Scientific Revolution, Chicago: University of Chicago Press.
  • Domski, M., 2012, “Locke's Qualified Embrace of Newton's Principia”, in Interpreting Newton: Critical Essays, ed. A. Janiak, and E. Schliesser, Cambridge: Cambridge University Press.
  • Downing, L., 1997, “Locke's Newtonianism and Lockean Newtonianism”, Perspectives on Science, 5(3): 285–310.
  • –––, 1998, “The Status of Mechanism in Locke's Essay,” The Philosophical Review, 107 (3): 381–414.
  • –––, 2007, “Locke's Ontology,” in L. Newman (ed.), The Cambridge Companion to Locke's Essay, Cambridge: Cambridge University Press, pp. 352–380.
  • Galileo, 1623, “The Assayer”, in M.R. Matthews (ed.), The Scientific Background to Modern Philosophy: Selected Readings, Indianapolis: Hackett, 1989.
  • Henry, J., 1994, “‘Pray do not ascribe that notion to me’: God and Newton's Gravity”, in J.E. Force and R. H. Popkin (eds.), The Books of Nature and Scripture: Recent Essays on Natural Philosophy, Theology and Biblical Criticism in the Netherlands of Spinoza's Time and the British Isles of Newton's Time, Dordrecht: Kluwer Academic Publishers, pp. 123–147.
  • Hill, J., 2004, “Locke's Account of Cohesion and its Philosophical Significance,” British Journal for the History of Philosophy, 12 (4): 611 – 630.
  • Janiak, A., 2008, Newton as Philosopher, Cambridge: Cambridge University Press.
  • Jardine, N., 1991, “Demonstration, Dialectic, and Rhetoric, in Galileo's Dialogue,” in D.R. Kelley and R.H.Popkin, Shapes of Knowledge from the Renaissance to the Enlightenment, Berlin: Springer, pp. 101–121.
  • Jolley, N., 2002, Locke: His Philosophical Thought, Oxford: Oxford University Press.
  • Kochiras, H. 2011. “Gravity's Cause and Substance Counting: Contextualizing the Problems.” Studies in History and Philosophy of Science, 42(1): 167–184.
  • Koyre, A., 1965, Newtonian Studies, London: Chapman & Hall.
  • Mandelbaum, M., 1964, Philosophy, Science, and Sense Perception, Baltimore: The Johns Hopkins Press.
  • McCann, E., 1994: “Locke's Philosophy of Body” in V. Chapell (ed.) The Cambridge Companion to Locke, Cambridge: Cambridge University Press, pp. 56–88.
  • –––, 2002, “John Locke”, in S. Nadler, A Companion to Early Modern Philosophy, Oxford: Blackwell Publishing.
  • McGuire, J.E., 1970, “Atoms and the Analogy of Nature,” reprinted in J.E. McGuire, Tradition and Innovation: Newton's Metaphysics of Nature (The University of Western Ontario Series in the Philosophy of Science), Dordrecht: Kluwer Academic, 1995.
  • Osler, M.J., 1998, “Mixing Metaphors: Science and religion or natural philosophy and theology in early modern Europe,” History of Science, 36: 91–113.
  • Osler, M.J., 1970, “John Locke and the Changing Ideal of Scientific Knowledge,” Journal of the History of Ideas, 31 (1): 3–16.
  • Palmieri, P., 1998 “Re-examining Galileo's Theory of Tides”, Archiv. Hist. Exact Sci. 53 (1998) 223–375.
  • –––, 2009, “A phenomenology of Galileo's experiments with pendulums”. British Journal for the History of Science, 42(4): 479–513, December 2009.
  • Park, K. and Daston, L., 2006, “Introduction: The Age of the New,” in K.Park and L. Daston (eds.), The Cambridge History of Science (Vol. 3, Early Modern Science), Cambridge: Cambridge University Press.
  • Rogers, G.A.J., 1982, “The System of Locke and Newton” in Z. Bechler (ed.), Contemporary Newtonian Research, Dordrecht: D. Reidel Publishing Co., pp. 215–238.
  • Schliesser, E., 2011, “Without God: Newton's Relational Theory of Attraction”, in D. Jalobeanu and P. Anstey (eds.), Vanishing Matter and the Laws of Motion: Descartes and Beyond, London: Routledge, pp. 80–100.
  • Smith, R., 2009, “Aristotle's Logic”, The Stanford Encyclopedia of Philosophy (Spring 2009 Edition), Edward N. Zalta (ed.), URL = <http://plato.stanford.edu/archives/spr2009/entries/aristotle-logic/>.
  • Stein, Howard, 2002, “Newton's Metaphysics”, in I. Bernard Cohen and George E. Smith (eds.), The Cambridge Companion to Newton, Cambridge University Press, 2002, pp. 256–307.
  • –––, 1993, “On Philosophy and Natural Philosophy,” Midwest Studies in Philosophy, 18 (1): 177–201.
  • –––, 1990, “Locke, the Great Huygenius, and the Incomparable Mr. Newton”, in P. Bricker, R. I. G. Hughes (eds.), Philosophical perspectives on Newtonian science, Cambridge, MA: MIT Press, pp. 17–48.
  • Stuart, M., 1998, “Locke on Superaddition and Mechanism,” British Journal for the History of Philosophy, 6 (3): 351–379.
  • Uzgalis, W., 2007, “John Locke”, The Stanford Encyclopedia of Philosophy (Summer, 2007 Edition), Edward N. Zalta (ed.), URL = <http://plato.stanford.edu/archives/sum2007/entries/locke/>.
  • Van Dyck, M., 2005, “The Paradox of Conceptual Novelty and Galileo's Use of Experiments”, Philosophy of Science, (72) 5: 864–875.
  • Westfall, R. S., 1980, Never at Rest: a Biography of Isaac Newton, New York: Cambridge University Press.
  • Wilson, M., 1991, “Superadded Properties: The Limits of Mechanism in Locke,” reprinted in Wilson, Ideas and Mechanism: Essays on Early Modern Philosophy, Princeton: Princeton University Press, pp. 196–214.
  • Wisan, W.L., 1978, “Galileo's scientific method: a reexamination”, in R.E. Butts and J.C. Pitt (eds.), New Perspectives on Galileo, Dordrecht: D. Reidel, pp. 1–58.
  • Winkler, Kenneth P., 2008, “Locke's Defense of Mathematical Physics”, Paul Hoffman, David Owen, and Gideon Yaffe (eds.), Contemporary Perspectives on Early Modern Science: Essays in Honor of Vere Chappell. Toronto: Broadview Press, 231–252.
  • Woolhouse, R.S., 1971, Locke's Philosophy of Science and Knowledge, Oxford: Blackwell.
  • –––, 1994, “Locke's Theory of Knowledge” in Vere Chappell (ed.), The Cambridge Companion to Locke, New York: Cambridge University Press, 146–171.
  • –––, 2005, “Locke and the Nature of Matter,” in C. Mercer and E. O'Neill (eds.), Early Modern Philosophy: Mind, Matter, and Metaphysics, Oxford: Oxford University Press, pp. 142–161.
  • Yolton, J., 1969, “The Science of Nature” in John W. Yolton (ed.), John Locke: Problems and Perspectives, Cambridge: Cambridge University Press, 183–193.

Academic Tools

sep man icon How to cite this entry.
sep man icon Preview the PDF version of this entry at the Friends of the SEP Society.
inpho icon Look up this entry topic at the Indiana Philosophy Ontology Project (InPhO).
phil papers icon Enhanced bibliography for this entry at PhilPapers, with links to its database.

Other Internet Resources

Related Entries

Boyle, Robert | demonstration: medieval theories of | Descartes, René | knowledge: analysis of | Locke, John | Newton, Isaac | Newton, Isaac: philosophy | primary and secondary qualities | rationalism vs. empiricism | sensation | substance

Copyright © 2013 by
Hylarie Kochiras <hk1@post.tau.ac.il>

Open access to the SEP is made possible by a world-wide funding initiative.
Please Read How You Can Help Keep the Encyclopedia Free